Tester
Lipp Laboratory, Department of Environmental Health Science, University of Georgia
2020-11-17
Load all required libraries.
library(tidyverse)## Warning: package 'tidyverse' was built under R version 3.6.3## -- Attaching packages ---------------------------------------------------------------------------- tidyverse 1.3.0 --## v ggplot2 3.3.2 v purrr 0.3.4
## v tibble 3.0.3 v dplyr 1.0.0
## v tidyr 1.1.0 v stringr 1.4.0
## v readr 1.3.1 v forcats 0.5.0## Warning: package 'ggplot2' was built under R version 3.6.3## Warning: package 'tibble' was built under R version 3.6.3## Warning: package 'readr' was built under R version 3.6.3## Warning: package 'dplyr' was built under R version 3.6.3## Warning: package 'forcats' was built under R version 3.6.3## -- Conflicts ------------------------------------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(plotly)## Warning: package 'plotly' was built under R version 3.6.3##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(broom)## Warning: package 'broom' was built under R version 3.6.3Read in raw data from RDS.
raw_data <- readRDS("./n1_n2_cleaned_cases.rds")Make a few small modifications to names and data for visualizations.
final_data <- raw_data %>% mutate(log_copy_per_L = log10(mean_copy_num_L)) %>%
rename(Facility = wrf) %>%
mutate(Facility = recode(Facility,
"NO" = "WRF A",
"MI" = "WRF B",
"CC" = "WRF C"))Seperate the data by gene target to ease layering in the final plot
#make three data layers
only_positives <<- subset(final_data, (!is.na(final_data$Facility)))
only_n1 <- subset(only_positives, target == "N1")
only_n2 <- subset(only_positives, target == "N2")
only_background <<-final_data %>%
select(c(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke)) %>%
group_by(date) %>% summarise_if(is.numeric, mean)
#specify fun colors
background_color <- "#7570B3"
seven_day_ave_color <- "#E6AB02"
marker_colors <- c("N1" = '#1B9E77',"N2" ='#D95F02')#remove facilty C for now
#only_n1 <- only_n1[!(only_n1$Facility == "WRF C"),]
#only_n2 <- only_n2[!(only_n2$Facility == "WRF C"),]
only_n1 <- only_n1[!(only_n1$Facility == "WRF A" & only_n1$date == "2020-11-02"), ]
only_n2 <- only_n2[!(only_n2$Facility == "WRF A" & only_n2$date == "2020-11-02"), ]Build the main plot
#first layer is the background epidemic curve
p1 <- only_background %>%
plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~new_cases_clarke,
type = "bar",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Daily Cases: ', new_cases_clarke),
alpha = 0.5,
name = "Daily Reported Cases",
color = background_color,
colors = background_color,
showlegend = FALSE) %>%
layout(yaxis = list(title = "Clarke County Daily Cases", showline=TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#renders the main plot layer two as seven day moving average
p1 <- p1 %>% plotly::add_trace(x = ~date, y = ~X7_day_ave_clarke,
type = "scatter",
mode = "lines",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Seven-Day Moving Average: ', X7_day_ave_clarke),
name = "Seven Day Moving Average Athens",
line = list(color = seven_day_ave_color),
showlegend = FALSE)
#renders the main plot layer three as positive target hits
p2 <- plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n1,
symbol = ~Facility,
marker = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n2,
symbol = ~Facility,
marker = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
layout(yaxis = list(title = "SARS CoV-2 Copies/L",
showline = TRUE,
type = "log",
dtick = 1,
automargin = TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#adds the limit of detection dashed line
p2 <- p2 %>% plotly::add_segments(x = as.Date("2020-03-14"),
xend = ~max(date + 10),
y = 3571.429, yend = 3571.429,
opacity = 0.35,
line = list(color = "black", dash = "dash")) %>%
layout(annotations = list(x = as.Date("2020-03-28"), y = 3.8, xref = "x", yref = "y",
text = "Limit of Detection", showarrow = FALSE))
p1## Warning: `arrange_()` is deprecated as of dplyr 0.7.0.
## Please use `arrange()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.## Warning: Ignoring 1 observations p2## Warning: `group_by_()` is deprecated as of dplyr 0.7.0.
## Please use `group_by()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.p_combined <-
plotly::subplot(p2,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationsp_combinedSave the plot to pull into the index
save(p_combined, file = "./plotly_fig.rda")Save an htmlwidget for website embedding
htmlwidgets::saveWidget(p_combined, "plotly_fig.html")Build loess smoothing figures figures
#create smoothing data frames
#n1
smooth_n1 <- only_n1 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N1")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#n2
smooth_n2 <- only_n2 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N2")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#add trendlines
#extract data from geom_smooth
#n1 extract
# *********************************span 0.6***********************************
#*****************Must always update the n = TOTAL NUMBER OF DAYS*************************
extract_n1 <- ggplot(smooth_n1, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n1<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 154)## Warning: Ignoring unknown aesthetics: outfit#n2 extract
extract_n2 <- ggplot(smooth_n2, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n2<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 154)## Warning: Ignoring unknown aesthetics: outfit#look at the fits to align dates and total observations
#n1
extract_n1## `geom_smooth()` using formula 'y ~ x'
fit_n1## [1] 11.58770 11.66753 11.74674 11.82456 11.90023 11.97298 12.04206 12.10670
## [9] 12.16756 12.22589 12.28181 12.33543 12.38687 12.43625 12.48368 12.52927
## [17] 12.57315 12.61543 12.65623 12.69566 12.73384 12.77088 12.80515 12.83514
## [25] 12.86119 12.88359 12.90266 12.91873 12.93210 12.94310 12.95204 12.95923
## [33] 12.96499 12.96964 12.97349 12.97686 12.98006 12.98341 12.98723 12.99183
## [41] 12.99753 13.00464 13.01348 13.02219 13.02888 13.03385 13.03740 13.03981
## [49] 13.04137 13.04238 13.04313 13.04390 13.04499 13.04670 13.04931 13.05312
## [57] 13.05841 13.06548 13.07462 13.08612 13.10028 13.11738 13.13771 13.16158
## [65] 13.19128 13.22835 13.27189 13.32100 13.37479 13.43236 13.49283 13.55528
## [73] 13.61884 13.68260 13.74567 13.80716 13.86616 13.92179 13.97314 14.01933
## [81] 14.05946 14.09264 14.11796 14.13454 14.14147 14.13820 14.12532 14.10370
## [89] 14.07421 14.03771 13.99506 13.94715 13.89482 13.83896 13.78041 13.72006
## [97] 13.65876 13.59739 13.53680 13.47787 13.42146 13.36843 13.31966 13.27601
## [105] 13.23834 13.20753 13.18153 13.15768 13.13589 13.11608 13.09816 13.08206
## [113] 13.06768 13.05494 13.04376 13.03406 13.02574 13.01873 13.01294 13.00829
## [121] 13.00469 13.00206 13.00031 12.99936 12.99913 12.99953 13.00048 13.00261
## [129] 13.00654 13.01216 13.01938 13.02807 13.03814 13.04948 13.06198 13.07554
## [137] 13.09005 13.10540 13.12148 13.13820 13.15544 13.17310 13.19125 13.21006
## [145] 13.22955 13.24975 13.27068 13.29237 13.31485 13.33814 13.36226 13.38725
## [153] 13.41312 13.43991#n2
extract_n2## `geom_smooth()` using formula 'y ~ x'
fit_n2## [1] 11.37120 11.49056 11.60835 11.72377 11.83601 11.94427 12.04774 12.14563
## [9] 12.23859 12.32794 12.41379 12.49627 12.57551 12.65164 12.72478 12.79506
## [17] 12.86260 12.92754 12.98999 13.05009 13.10797 13.16374 13.21575 13.26246
## [25] 13.30419 13.34126 13.37400 13.40272 13.42774 13.44939 13.46799 13.48385
## [33] 13.49731 13.50867 13.51825 13.52639 13.53340 13.53960 13.54532 13.55087
## [41] 13.55657 13.56274 13.56972 13.57448 13.57418 13.56936 13.56058 13.54840
## [49] 13.53336 13.51601 13.49691 13.47661 13.45566 13.43461 13.41402 13.39443
## [57] 13.37640 13.36047 13.34721 13.33716 13.33088 13.32891 13.33181 13.34013
## [65] 13.35450 13.37472 13.40015 13.43012 13.46399 13.50110 13.54078 13.58240
## [73] 13.62528 13.66878 13.71225 13.75501 13.79643 13.83583 13.87258 13.90601
## [81] 13.93546 13.96029 13.97983 13.99343 14.00044 14.00150 13.99794 13.99011
## [89] 13.97836 13.96304 13.94450 13.92310 13.89918 13.87309 13.84520 13.81585
## [97] 13.78538 13.75416 13.72253 13.69084 13.65946 13.62871 13.59897 13.57058
## [105] 13.54389 13.51925 13.49603 13.47331 13.45100 13.42902 13.40729 13.38571
## [113] 13.36422 13.34271 13.32111 13.29934 13.27731 13.25493 13.23213 13.20881
## [121] 13.18490 13.16030 13.13495 13.10874 13.08160 13.05344 13.02419 12.99443
## [129] 12.96476 12.93505 12.90517 12.87500 12.84441 12.81329 12.78150 12.74893
## [137] 12.71545 12.68093 12.64526 12.60830 12.56994 12.53005 12.48883 12.44659
## [145] 12.40336 12.35917 12.31404 12.26801 12.22111 12.17337 12.12481 12.07548
## [153] 12.02539 11.97458#assign fits to a vector
n1_trend <- fit_n1
n2_trend <- fit_n2
#extract y min and max for each
limits_n1 <- ggplot_build(extract_n1)$data## `geom_smooth()` using formula 'y ~ x'limits_n1 <- as.data.frame(limits_n1)
n1_ymin <- limits_n1$ymin
n1_ymax <- limits_n1$ymax
limits_n2 <- ggplot_build(extract_n2)$data## `geom_smooth()` using formula 'y ~ x'limits_n2 <- as.data.frame(limits_n2)
n2_ymin <- limits_n2$ymin
n2_ymax <- limits_n2$ymax
#reassign dataframes (just to be safe)
work_n1 <- smooth_n1
work_n2 <- smooth_n2
#fill in missing dates to smooth fits
work_n1 <- work_n1 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n1 <- work_n1$date
work_n2 <- work_n2 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n2 <- work_n2$date
#create a new smooth dataframe to layer
smooth_frame_n1 <- data.frame(date_vec_n1, n1_trend, n1_ymin, n1_ymax)
smooth_frame_n2 <- data.frame(date_vec_n2, n2_trend, n2_ymin, n2_ymax)#make plotlys
#plot smooth frames
p3 <- plotly::plot_ly() %>%
plotly::add_lines(x = ~date_vec_n1, y = ~n1_trend,
data = smooth_frame_n1,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1,
'</br> Median Log Copies: ', round(n1_trend, digits = 2),
'</br> Target: N1'),
line = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_lines(x = ~date_vec_n2, y = ~n2_trend,
data = smooth_frame_n2,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2,
'</br> Median Log Copies: ', round(n2_trend, digits = 2),
'</br> Target: N2'),
line = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_ribbons(x ~date_vec_n1, ymin = ~n1_ymin, ymax = ~n1_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1, #leaving in case we want to change
'</br> Max Log Copies: ', round(n1_ymax, digits = 2),
'</br> Min Log Copies: ', round(n1_ymin, digits = 2),
'</br> Target: N1'),
name = "",
line = list(color = '#1B9E77')) %>%
plotly::add_ribbons(x ~date_vec_n2, ymin = ~n2_ymin, ymax = ~n2_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2, #leaving in case we want to change
'</br> Max Log Copies: ', round(n2_ymax, digits = 2),
'</br> Min Log Copies: ', round(n2_ymin, digits = 2),
'</br> Target: N2'),
name = "",
line = list(color = '#D95F02')) %>%
layout(yaxis = list(title = "Total Log SARS CoV-2 Copies",
showline = TRUE,
automargin = TRUE)) %>%
layout(xaxis = list(title = "Date")) %>%
plotly::add_segments(x = as.Date("2020-06-24"),
xend = as.Date("2020-06-24"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Bars Repoen",
hoverinfo = "text",
text = "</br> Bars Reopen",
"</br> 2020-06-24",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-07-09"),
xend = as.Date("2020-07-09"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Mask Mandate",
hoverinfo = "text",
text = "</br> Mask Mandate",
"</br> 2020-07-09",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-08-20"),
xend = as.Date("2020-08-20"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> Classes Begin",
"</br> 2020-08-20",
hoverinfo = "text",
text = "Classes Begin",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-10-03"),
xend = as.Date("2020-10-03"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> First Home Football Game",
"</br> 2020-10-03",
hoverinfo = "text",
text = "First Home Football Game",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n1,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#1B9E77', size = 6, opacity = 0.65)) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n2,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#D95F02', size = 6, opacity = 0.65))
p3Create final trend plot by stacking with epidemic curve
smooth_extract <-
plotly::subplot(p3,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationssmooth_extractsave(smooth_extract, file = "./smooth_extract.rda")